Nuclearn Document Library Book a demo
NEW Nuclearn Document Library and Nuclearn Connect

Your plant's documents,
finally in one place.

Nuclearn Connect brings procedures, vendor manuals, drawings and licensing documents in from SharePoint, file shares and your document systems. Scanned pages are read with OCR, so old documents are searchable too. And every AtomAssist answer traces back to the page, the passage and the revision it came from.

Book a demo See what it does
  • Works from the systems you already have
  • Connect only connects out
  • People see only what they may read
AtomAssist answering from a project of NRC Regulatory Guides: the library on the left, an answer about Regulatory Guide 1.192 Revision 5 with numbered citations in the middle, and a Sources panel listing each cited passage with its document, page and revision.

Screenshots on this page show publicly available NRC documents, as they appear in the Nuclearn platform today.

  • Backed by NuclearnAI software built for nuclear plants since 2021
  • Over 70 facilitiesNuclearn products serve sites in North America and the U.K.
  • ISO/IEC 27001:2022Certified information security management
Why teams use it

One library for the jobs that start with finding a document.

Work planning, engineering questions, licensing research and turnover all begin the same way: find the right document, and make sure it's the current one. The library does that across your document systems.

Work planning

Gather what a work package needs

Find past work packages, vendor manuals, procedures and engineering changes in one search, then collect them in a project folder for the job.

See search → Everyone

Search the scans, too

Typed guides, old work packages and vendor manuals that exist only as scans are read with OCR, then found by search and cited in answers like any other page.

See OCR → Engineering and licensing

Trace an answer to its source

Every AtomAssist answer cites its sources. Follow a citation to the page, the highlighted passage and the document's revision, and export the trail as a report.

See traceability → Everyone

Know which revision you're reading

Each document keeps its revision history. Open an older revision and a banner names the current one; project folders flag documents with a newer revision.

See revisions → Document control and IT

Keep the library in step with your systems

Nuclearn Connect syncs SharePoint, file shares, CMIS repositories, Maximo attachments and Aras Innovator. Changes and deletions follow the source.

See Nuclearn Connect → IT and security

Search only what each person may read

Search, listings and AtomAssist answers reach only the collections and project folders a person may read. If they can read nothing, search returns nothing.

See access control →
The problem

The documents exist. Finding the right one takes the time.

A plant's documents live in several places: a document management system, SharePoint sites, file shares, Maximo attachments, a PLM system. Planners and engineers search each one, then check by hand that the copy they found is the current revision.

INL's work-management study describes the same thing: technical documentation "may be in various repositories and formats at the power plant", and "current revisions of the procedures must be included" in every work package.1

The Document Library starts from the systems you already have.

$1.6M1
per reactor unit, per year

INL's estimate of what four work-package planning steps cost today. Each starts with a document search: similar past work packages, the technical information for the equipment, engineering changes and technical specifications, and the current revisions of procedures. Online work only, at $200/hr.

15 min1
on each work package

INL's estimate of the time a planner spends searching for engineering changes and technical specification changes that affect the equipment, then comparing them with the package.

INL's figures are its estimates of what these steps cost now, not savings from Nuclearn. Its step times are expert estimates and vary by plant. Outage work is not included. Sources at the foot of the page.

How it works

From your document systems to a cited answer

No migration and no new system of record. Your documents stay where they are; the library keeps a searchable copy in step with them.

Connect your sources

Run Nuclearn Connect next to SharePoint, a file share or your document system and pick what to sync. Or upload files into a project folder.

SharePointFile shareCMISMaximoArasUpload

The library reads them

Each document is read, with OCR for scanned pages, and indexed for search. Changes in the source come through as they happen or on a schedule you set.

Search, ask, cite

Search the library, open any document in the browser, or ask AtomAssist. Answers cite the documents they used.

OCR for scanned documents

Scanned pages become text you can search and cite.

Much of a plant's history exists only as scans: typed guides, old work packages, vendor manuals. As each PDF comes in, the library finds the pages with no usable text and reads them with OCR. From then on, those pages turn up in search and in AtomAssist answers like any other page, and a citation opens the scan itself.

A library search for monthly operating report contents finds a 1974 Atomic Energy Commission regulatory guide that exists only as a scan. It opens at page 3, with the text read from the scan listed beside the page image. An AtomAssist citation opened on the same 1974 scan: the cited and read passages outlined and numbered on the scanned page.
A 1974 guide that exists only as a scan, found by a library search (left) and cited in an AtomAssist answer, with the passages outlined on the scanned page (right).
How it reads them

Found page by page, with the status in view.

Every PDF is checked when it arrives. Pages without a text layer are read with OCR, and so are PDFs whose text layer is there but unreadable. Text read from a scan is indexed with everything else.

  • Catches scanned pages inside an otherwise digital PDF.
  • Citations and highlights land on the scanned page itself.
  • Each document shows its stage while it's read, run through OCR and indexed, and can be reprocessed from its page.
A project folder of NRC Regulatory Guides showing 2 processing at the top, with the two scanned 1970s guides marked ocr document while they are read.
Traceability

Traceability from every answer back to its source.

An answer is only useful if someone can check it. Each link from an AtomAssist answer back to the document is a click away, and the whole trail can be exported.

  1. The answerEach statement carries a numbered citation.
  2. The citationNames the document, the page and the revision, and quotes the passage.
  3. The pageThe source opens at that page, with the cited passage highlighted.
  4. The revisionCurrent or superseded, with the document's revision history.
  5. The sourceA synced document's collection is fed by one source folder, and Connect shows when it last synced.

Every link follows access. Search, answers and citations draw only on documents the person may read.

Cited answers

Answers you can check in one click.

Start a chat from a document or a project folder and ask AtomAssist. Each answer lists what it cited, what it read and what it searched. Click a citation and the source opens beside the chat at the cited page, with the passage highlighted.

  • Chat with one document, or with a whole project folder.
  • Cited and read passages numbered and highlighted on the page.
  • Export the trail as a PDF source report, for one answer or the whole conversation: the question, who asked and when, the model, and each cited document's revision, page and passage. A JSON audit trail is there too.
An AtomAssist answer about Regulatory Guide 1.192 Revision 5, with numbered citations after each statement and a table of pump-related code cases. A citation opened beside the chat: page 4 of Regulatory Guide 1.192 Revision 5, with cited passages one to three outlined and labelled.
Revisions

See the history, and know when there's a newer one.

Upload a new version of a document and it becomes the current revision once it's read, with the earlier revisions kept in its history. Open an older revision and a banner says which revision replaced it and when. Project folders that use the document flag the newer revision and bring it in with a click.

  • Revision history: label, current or superseded, dates, and each revision's file.
  • Citations name the revision they came from.
  • Who brought a newer revision into a project, and when, is recorded.
Regulatory Guide 1.192 with its Revision History open: Rev 5 marked current, Rev 4 and an earlier revision marked superseded, each with created and superseded dates and buttons to view or download it.
Nuclearn Connect

Your document systems, synced from inside your network.

Nuclearn Connect runs next to your sources, behind your firewall. It reads the sources you choose and sends new and changed documents to your Nuclearn platform over outbound HTTPS. The platform never opens a connection into your network.

How Nuclearn Connect brings documents into the platform Inside the plant network, Nuclearn Connect reads SharePoint, file shares, CMIS repositories, Maximo and Aras Innovator. It sends files and changes out through the firewall over outbound HTTPS to the Nuclearn platform, where the Document Library, search and AtomAssist use them. Nothing connects inward. INSIDE YOUR PLANT NETWORK SharePoint Sites and libraries, via Microsoft Graph File shares Local disk, NFS or SMB mounts CMIS repositories Alfresco and other CMIS servers Maximo Attached documents Aras Innovator PLM file attachments Nuclearn Connect Runs next to your sources Reads the sources you pick Sends only what changed Signs in with its own key Opens no inbound ports Outbound HTTPS files and changes Your firewall Your Nuclearn platform Where people search, read and ask Document Library Collections, folders and revision history Search Limited to what each person may read AtomAssist Answers that cite the page they came from How Nuclearn Connect brings documents into the platform Inside the plant network, Nuclearn Connect reads SharePoint, file shares, CMIS repositories, Maximo and Aras Innovator. It sends files and changes out through the firewall over outbound HTTPS to the Nuclearn platform, where the Document Library, search and AtomAssist use them. Nothing connects inward. INSIDE YOUR PLANT NETWORK SharePoint Microsoft Graph File shares Local, NFS or SMB CMIS Alfresco and others Maximo Attached documents Aras Innovator PLM file attachments reads Nuclearn Connect Runs next to your sources Reads the sources you pick Sends only what changed Signs in with its own key Opens no inbound ports Outbound HTTPS files and changes Your firewall Your Nuclearn platform Where people search, read and ask Document Library Collections, folders, revisions Search Limited to what each person may read AtomAssist Answers that cite their sources
Connect makes every connection itself, outward. Even a request to browse your folders is something Connect asks the platform for.
SharePointSharePoint Online sites and document libraries, through Microsoft Graph.
File sharesAny folder Connect can mount: local disk, NFS or SMB. Picks up changes as files are saved.
CMIS repositoriesAlfresco and other CMIS 1.0 and 1.1 servers. Follows the repository's change log where it has one.
MaximoDocuments attached to Maximo records, through the Manage REST API.
Aras InnovatorFiles attached to documents, parts and change orders, at their current generation.
Keeping in step

When a document changes, the library follows.

Connect keeps checking what you chose to sync and sends only what changed. For file shares, SharePoint and CMIS you pick folders from a live view of the source; Maximo and Aras sync as a whole. The admin page shows each folder's status and when Connect last checked in.

How a change in a source reaches the library Step 1: a document is added, edited or deleted in the source. Step 2: Connect notices, as files are saved on file shares, through a frequent change check on CMIS repositories and Aras, or on a scheduled check for other sources. Step 3: only the change is sent; new and edited files are uploaded and deletions are applied. Step 4: the file is read and indexed again, so search and AtomAssist use the new copy. Then Connect waits for the next change. 1 Something changes A document is added, editedor deleted in SharePoint, afile share or another source. 2 Connect notices As files are saved on a share,by a frequent change check onCMIS and Aras, or on a schedule. 3 Only the change goes up New and edited files are sent.Deletions are applied, with abrake on large batches. 4 The library is current The file is read and indexedagain, so search and AtomAssistanswer from the new copy. then waits for the next change How a change in a source reaches the library Step 1: a document is added, edited or deleted in the source. Step 2: Connect notices, as files are saved on file shares, through a frequent change check on CMIS repositories and Aras, or on a scheduled check for other sources. Step 3: only the change is sent; new and edited files are uploaded and deletions are applied. Step 4: the file is read and indexed again, so search and AtomAssist use the new copy. 1 Something changes A document is added, edited or deleted inSharePoint, a file share or another source. 2 Connect notices As files are saved on a share, by a frequentchange check on CMIS and Aras, or on a schedule. 3 Only the change goes up New and edited files are sent. Deletions areapplied, with a brake on large batches. 4 The library is current The file is read and indexed again, so searchand AtomAssist answer from the new copy. Then Connect waits for the next change.
Built for plant documents

Made for decades of documents, not one tidy system

Metadata read for you

Title, author, document type, date and version are read from each document by AI, shown next to the record's own facts and marked as AI-generated.

Upload too

Drag files into a project folder: PDF, Word, Excel, PowerPoint, text and more.

Notes on the page

People who can edit a document highlight passages and add notes on its page.

Folders for each job

Gather documents for an outage, a modification or a licensing action in a shared project folder.

Synced stays synced

Synced collections can't be edited by hand, so the library matches the source.

Shared the way your teams work

Share a project with people or groups to view, edit or manage. Keep a collection private, share it with a team, or open it to your whole organization. Adding a document to a project doesn't change who can read it.

Access and security

Connected from inside. Read only by the right people.

Connect needs no inbound access to your network, and the library shows each person only what they're allowed to read.

Outbound only

Connect runs inside your network and makes every connection itself, over HTTPS. It opens no listening ports, and the platform never connects in.

A key for each Connect

Each Connect gets its own key, shown once; the platform keeps only a hash. Rotate or revoke it from the admin page. TLS certificates are checked, and Connect can trust your own certificate authority.

Read-only on your sources

Connect only reads from your document systems. It doesn't write, move or delete anything there.

Private until shared

A collection created by a sync starts private to the person who set it up. Access is then given per collection, to people or groups.

Search follows access

Search, listings and AtomAssist answers reach only what the person may read. Admins manage access but don't automatically read private collections.

A certified company

Nuclearn is ISO/IEC 27001:2022 certified. Sign-in goes through the Nuclearn platform and its single sign-on.

Coming next

Revisions found for you, and what changed between them.

In development now, and not yet in a release.

Document control

Revisions linked automatically

Read the document number and revision from each file, link new revisions to the document they replace, and queue uncertain matches for a person to confirm.

Everyone

What changed

A section-by-section summary of what was added, removed and changed between two revisions, next to the passages themselves.

Engineering

Revisions and details from your sources

Released versions from a PLM or document system kept as revisions, with the source's item number, revision and lifecycle state shown on each record.

Document control and IT

Documentum and FileNet

Dedicated Connect adapters for OpenText Documentum and IBM FileNet, with released-version filtering.

What it is not

The Document Library is not a document control system or a records vault, and Connect doesn't write to your sources. Your document systems stay the system of record; the library is the searchable copy people read and ask questions of.

FAQ

Questions plants ask first

Do we have to move our documents?

No. Your documents stay in SharePoint, your file shares and your document systems. Connect keeps a copy in the library in step with them, and synced collections can't be edited by hand, so the source stays the source.

Which systems can Connect sync today?

SharePoint (through Microsoft Graph), file shares (local disk, NFS or SMB), CMIS 1.0 and 1.1 repositories such as Alfresco, documents attached in Maximo, and Aras Innovator. Dedicated Documentum and FileNet adapters are in development. Anything else can be uploaded.

Does Connect need a firewall opening?

No inbound opening. Connect runs inside your network and makes outbound HTTPS calls to your Nuclearn platform. It opens no listening ports; even browsing your folders from the platform works by Connect asking for the request.

Are our source permissions copied?

Not document by document. Access is set per collection in Nuclearn: a synced collection starts private, and you share it with the people and groups who should read it. Search and AtomAssist then follow those permissions.

What happens when a document is deleted or moved in the source?

Deletions are applied to the library. A brake holds back unusually large batches of deletions, or deletions after an incomplete check of the source, until an admin confirms them. A moved document is removed from its old place and added at the new one.

Do synced documents keep their earlier revisions?

Not yet. Connect keeps the library's copy current, so a changed file replaces the earlier copy. Revision history today comes from uploading a new version in the library. Keeping released versions from your source systems as revisions is in development.

Which scanned documents can it read?

PDFs with scanned or image-only pages, including the odd scanned page inside an otherwise digital PDF, and PDFs whose text layer is there but unreadable. The library checks each PDF when it arrives and runs OCR where it's needed. A collection can also have the pictures and figures in its PDFs described in words for search; an administrator turns that on per collection.

Can AtomAssist's answers be checked, and is there a record?

Yes. Each answer lists the sources it cited and the ones it read, and a citation opens the source at the cited page with the passage highlighted. Check the passage before you rely on the answer. For a record, export a PDF source report or a JSON audit trail for one answer or the whole conversation, or save the report into a project folder.

Is the Document Library part of AtomAssist?

Yes. The Library sits in AtomAssist Projects, next to the chat. Nuclearn Connect is added to a deployment and set up by an administrator.

Book a demo

See a question answered from your document systems.

We'll show Connect syncing a document source, the library and its search, and AtomAssist answering with citations that open the page, on public NRC documents. We'll also answer your IT team's questions about Connect, sign-in and access.

  • A live walkthrough on publicly available NRC documents
  • Which of your document systems Connect can sync
  • Your questions on deployment, sign-in and access

The Nuclearn Document Library is backed by Nuclearn, whose products serve over 70 facilities in North America and the U.K.

Book a demo

Tell us a little about your site and we'll set up a time.

Prefer email? contact@nuclearn.ai